The Standings → GPT‑6 Luna (high)
OpenAI
high reasoning effort · via Codex CLI · judged by Claude Sonnet 5 · the setting shown on the standings; also run at xhigh49.7 medium47.3
Results from run ‘first’: ten items, ten verdicts.
GPT‑6 Luna (high) scores 54 on craft, 9 points below the field’s 63 and 39 points short of the authors themselves. Its shape differs from the field’s: relatively strong on restraint, relatively weak on period detail and momentum. On the mechanical checks it kept to the brief’s word range on 9 of 10 items.
The record: 10 items · 10 verdicts · 0 errors · 2,068 words on average · est. generation cost $0.04
Eight dimensions, each 0–100, averaged over ten items; the author’s real chapter was in the room every time.
Craft by dimension, 0–100 · judged by Sonnet 5 beside the author’s real chapter
The authors themselves average 93 on the five writing dimensions, the field 63, this model 54. Hover or focus a bar for the field and the authors on that dimension.
The judge’s one-sentence best quality and worst flaw for each chapter, written for a general reader.
Best Renfield's flat, repeated question about Miss Westenra by name, with no explanation offered, is genuinely unsettling and well-placed.
Worst The closing telegram says Jonathan is already safely home by early June, which flatly contradicts the source material's own timeline, where he is still trapped abroad for weeks after that.
Best It hits its four required story beats cleanly and even manages one nice original touch — the stopped clock reading 'five minutes to five,' which quietly dovetails with the time of death established in the source chapter.
Worst The people barely come alive — Alfred, John, and Cynthia deliver their lines with little of the nervous, guarded texture real suspects under suspicion would show, so the scene reads more like a checklist of interviews than a lived morning.
Best It hits every required plot point cleanly and in a sensible order, so a reader never loses track of what is supposed to be happening.
Worst Edna receives a letter from Léonce with the exact same news in the afternoon and then a telegram repeating it at dinner, which reads like an oversight rather than a choice.
Best It hits every required plot point cleanly and lands on exactly the ending state the assignment calls for.
Worst The dialogue reads like a modern script with short, flat back-and-forth lines rather than Jane Austen's actual sentence style.
Best It hits the required story beats cleanly and directly, and lands the ending exactly where the assignment asked it to.
Worst The prose often explains what characters think and feel instead of showing it, and the key burn scene doesn't logically hold together, since a stick touches the object but the man's hand is what gets burned.
Best Lord Emsworth's pince-nez getting caught behind his collar again is a nice, in-character touch of his established forgetfulness, and Baxter's officious grilling about the Dewey system gives the interview a real obstacle.
Worst The very last scene is confusing: the umbrella is described as already sitting beside Psmith, but then the porter walks in as if delivering it for the first time, so the small payoff moment doesn't quite land.
Best It's clean and easy to follow, and it lands on the exact required closing image without fuss.
Worst It reads like a competent summary of the required plot points rather than Betteredge talking to us, with flat dialogue and none of his chatty asides to the reader.
Best The scene with the wealthy rival Gerald is clearly staged and easy to follow, hitting the required story beat without confusion.
Worst It spells out Jake's jealousy in a flat explanatory paragraph near the end instead of letting the reader feel it, which breaks the show-don't-tell style the whole book depends on.
Best The anonymous typed warning note about the check is a nicely understated, quietly threatening touch for a gangster who prefers not to show his face.
Worst The widow's interview is a long string of short denials ('I won't tell you') that reads like a deposition rather than a scene, and it never explains how blood got on her shoe after she insists she never went near the body, leaving the chapter's own central clue hanging.
Best It moves through every required event in a clear, easy-to-follow order without ever losing the thread of the journey.
Worst Most sentences just report what happened ('There was very little to see outside') instead of making the reader feel the cold, the dark, or the strangeness of the house, so the chapter reads like a summary rather than a scene.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| Dracula item 015 · after ch. 5 · Bram Stoker | 58.0 | 95 | 61 | 1,937 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 53.0 | 95 | 56 | 2,069 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 52.1 | 95 | 55 | 1,440 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 51.7 | 95 | 54 | 2,440 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 50.4 | 90 | 56 | 2,124 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 49.6 | 95 | 52 | 2,469 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 48.8 | 90 | 54 | 2,428 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 47.7 | 95 | 50 | 1,648 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 47.5 | 95 | 50 | 2,191 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 47.1 | 90 | 52 | 1,930 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Nine of ten items landed inside their brief’s word range.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The widow's interview is a long string of short denials ('I won't tell you') that reads like a deposition rather than a scene, and it never explains how blood got on her shoe after she insists she never went near the body, leaving the chapter's own central clue hanging.
The people barely come alive — Alfred, John, and Cynthia deliver their lines with little of the nervous, guarded texture real suspects under suspicion would show, so the scene reads more like a checklist of interviews than a lived morning.
Most sentences just report what happened ('There was very little to see outside') instead of making the reader feel the cold, the dark, or the strangeness of the house, so the chapter reads like a summary rather than a scene.
The very last scene is confusing: the umbrella is described as already sitting beside Psmith, but then the porter walks in as if delivering it for the first time, so the small payoff moment doesn't quite land.
Automated screening results across the ten outputs.